Februaey 16, 1906.] 



SCIENCE. 



249 



quired a yellowish color which serves to 

 distinguish it from all other members of 

 the group. 



Ammospermophilus harrisi, inhabiting 

 the southwestern half of Arizona and ex- 

 tending south far into the state of Sonora ; 

 separated from its neighbor {leucurus and 



|gSS^ h.aix.ii!i. 

 Mr^ iatprpfftg. 



Fig. 1. Distribution of Ground Squirrels of the 

 Genus Ammospermophilus. 



subspecies) on the west, north and east by 

 two kinds of barriers — on the west by the 

 absolute barrier of the Colorado River ; on 

 the north and east, along the southern edge 

 of the Arizona plateau and its southerly 

 continuation in New and Old Mexico, by 

 the climatic barrier interposed by the ele- 

 vation of the land. 



Ammospermophilus leucimis, inhabiting 

 the Colorado and Mohave deserts in Cali- 

 fornia and thence, passing north of harrisi, 

 ranging northerly and easterly over the 

 Sonoran deserts of the Great Basin and 

 easterly over northern Arizona to the Paint- 

 ed Desert and the Puerco, where it changes 

 into subspecies cinnamomeus, which con- 

 tinues easterly to the Rio Grande Valley 



near the center of New Mexico. On the 

 opposite side of the Rio Grande (east of 

 Albuquerque) another subspecies, inter- 

 pres, begins and follows the Rio Grande 

 southerly and easterly as far as the valley 

 of the Pecos. A third subspecies, penin- 

 sula, occurs in the peninsula of Lower Cali- 

 fornia. It thus appears that leucurus, in- 

 cluding the subspecies, surrounds harrisi 

 on three sides— west, north and east— form- 

 ing a complete horseshoe open only at the 

 south. The nature of the barriers sepa- 

 rating it from harrisi has been mentioned 

 under the latter species. 



Of the three subspecies of leucurus, two , 

 — cinnamomeus and peninsuloB — have the 

 under side of the tail white and are only 

 moderately accentuated color forms, con- 

 tinuoiis in distribution with the parent 

 form and intergrading with it by imper- 

 ceptible steps. The third subspecies, in- 

 terpres, has some black intermixed in the 

 white of the tail, and specimens from the 

 El Paso region, where geographically in- 

 terpres approaches nearest to harrisi, have 

 more black than specimens from near Al- 

 buquerque, suggesting the possibility that 

 intergradation may take place in the in- 

 tervening area, in southwestern New Mex- 

 ico. Unfortunately we have as yet no 

 specimens from this area. If intergrada- 

 tion should be shown to occur, then we 

 shall have an example of two well-defined 

 species occupying adjacent areas and in- 

 tergrading at the point of contact while 

 remaining distinct everywhere else. In 

 other words, the species are perfectly dis- 

 tinct wherever barriers exist separating 

 their ranges, and intergrade when there are 

 no barriers. In distribution and behavior, 

 therefore, these animals conform to the 

 usual rule, that closely related species sepa- 

 rated by impassable barriers remain dis- 

 tinct, while those not separated by such 

 barriers intergrade. 



Biit the most significant and complicated 



